EP3239925A1 - Fish type determination device and fish type determination method - Google Patents
Fish type determination device and fish type determination method Download PDFInfo
- Publication number
- EP3239925A1 EP3239925A1 EP14875888.1A EP14875888A EP3239925A1 EP 3239925 A1 EP3239925 A1 EP 3239925A1 EP 14875888 A EP14875888 A EP 14875888A EP 3239925 A1 EP3239925 A1 EP 3239925A1
- Authority
- EP
- European Patent Office
- Prior art keywords
- fish body
- image
- color
- fish
- monochrome
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Withdrawn
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/10—Image acquisition
- G06V10/12—Details of acquisition arrangements; Constructional details thereof
- G06V10/14—Optical characteristics of the device performing the acquisition or on the illumination arrangements
- G06V10/143—Sensing or illuminating at different wavelengths
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/255—Detecting or recognising potential candidate objects based on visual cues, e.g. shapes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/56—Extraction of image or video features relating to colour
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/80—Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level
- G06V10/803—Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level of input or preprocessed data
Definitions
- the present invention relates to a fish species identification device and a fish species identification method for identification a fish species.
- a person looks at a fish body and identifies a fish species thereof, and classifies it into the fish species according to an identification result.
- the fish species screening is performed manually within a limited time from landing to auction; however, since the fish species screening takes time, screening work is not completed within a desired time in some cases. Also, since the fish body is touched with a person's hands, there has been pointed out a problem in terms of freshness deterioration and hygiene.
- an automated fish species identification processing device has been proposed as a device for identifying a species of the fish body (see, for example, Patent Literature 1).
- the fish species identification processing device is provided with a reflected light imaging mechanism and a transmitted light imagingmechanism, and it performs fish species identification first by recognizing a surface pattern and a pectoral fin shape of the fish body by the reflected light imaging mechanism and then by detecting a shape of the fish body by the transmitted light imaging mechanism.
- the fish body which is placed on a conveyer, is conveyed to the reflected light imaging mechanism and to the transmitted light imaging mechanism in order.
- an ordinary conveyer that is black or the like is used as the conveyer, there is hardly any difference in luminance contrast between the conveyer and the fish body placed thereon, whereby it is not possible to accurately extract an outline of the fish body.
- Patent Literature 1 JP 10-111922 A
- the linear light filter is specially required.
- the transparent conveyer is also required, whereby a special measure for removing dirt due to the fish body is necessary.
- the transmitted light imaging mechanism it is necessary to radiate the irradiation light from beneath the conveyer, whereby there is a problem in that an irradiation mechanismbecomes complicated.
- the outline of the fish body is obtained in a light-transmitted form, bleeding of the light occurs in an outline portion, which may have an influence on determination of a shape and a size of the fish body.
- An objective of the present invention is to provide a simply-configured fish species identification device and a fish species identification method capable of solving various problems described above.
- a fish species identification device of the invention includes: a conveying means configured to convey a fish body in a fixed direction; a first irradiation means configured to irradiate the fish body with monochromatic light during conveying by the conveying means; a second irradiation means configured to irradiate the fish body with white light during the conveying by the conveying means in timing different from irradiation with the monochromatic light by the first irradiation means; a first imaging means configured to image a monochrome image including the fish body irradiated with the monochromatic light; a second imaging means configured to image a color image including the fish body irradiated with the white light; a feature quantity extraction means configured to extract a feature quantity of the fish body, by obtaining an outline of the fish body from a monochrome fish body image within the monochrome image and by detecting a measurement region of a color fish body image within the color image based on the outline that has been obtained,
- the invention of claim 2 is the fish species identification device according to claim 1, wherein the monochromatic light is a near-infrared light.
- the invention of claim 3 is the fish species identification device according to claim 1 or 2, wherein the feature quantity extraction means performs processing of detecting the measurement region of the color fish body image by overlapping the color fish body image on the monochrome fish body image.
- the invention of claim 4 is the fish species identification device according to claim 3, wherein the first imaging means and the second imaging means are constituted of a common imaging means, and the feature quantity extraction means performs the processing of detecting the measurement region of the color fish body image by shifting the color fish body image according to a time difference between imaging timing of the first imaging means and imaging timing of the second imaging means and by overlapping the color fish body image on the monochrome fish body image.
- the invention of claim 5 is the fish species identification device according to claim 3, wherein the first imaging means and the second imaging means are arranged side by side in a conveyance direction of the conveying means, a time difference between imaging timing of the first imaging means and imaging timing of the second imagingmeans is set according to a conveyance speed of the conveying means, and the feature quantity extraction means performs the processing of detecting the measurement region of the color fish body image by overlapping the color fish body image directly on the monochrome fish body image.
- a fish species identification method of the invention includes: conveying a fish body in a fixed direction; imaging a monochrome image including the fish body by irradiating the fish body with monochromatic light during the conveying; imaging a color image including the fish body by irradiating the fish body with white light during the conveying in timing different from irradiation with the monochromatic light; obtaining an outline of the fish body from a monochrome fish body image within the monochrome image; detecting a measurement region of a color fish body image within the color image based on the outline of the fish body; extracting a feature quantity of the fish body from a color pattern of the color fish body image within the outline and the measurement region; and by checking the feature quantity of the fish body that has been extracted with a feature quantity of one or more than one registered fish body registered in advance, identifying the feature quantity of which registered fish body is the closest to the feature quantity of the fish body.
- the invention of claim 7 is the fish species identification method according to claim 6, wherein the monochromatic light ray is a near-infrared light.
- the invention of claim 8 is the fish species identification method according to claim 6 or claim 7, wherein detecting of the measurement region of the color fish body image is performed by overlapping the color fish body image on the monochrome fish body image.
- the invention of claim 9 is the fish species identification method according to claim 8, wherein the monochrome image and the color image are obtained by one imaging means, and detecting of the measurement region of the color fish body image is performed by shifting the color fish body image according to a time difference between imaging timing of the monochrome image and imaging timing of the color image and by overlapping the color fish body image on the monochrome fish body image.
- the invention of claim 10 is the fish species identification method according to claim 8, wherein the monochrome image and the color image are obtained at a different position in the conveyance direction, a time difference between imaging timing of the monochrome image and imaging timing of the color image is set according to the conveyance speed, and detecting of the measurement region of the color fish body image is performed by overlapping the color fish body image directly on the monochrome fish body image.
- a conveyer of any color not limited to a transparent one may be used as a conveying means, and a special measure for removing dirt due to the fish body is not necessary.
- light can be radiated from above the conveying means, it is possible to simplify a configuration of an irradiation means compared to where the light is radiated frombeneath the conveyer.
- the outline of the fish body is obtained from a monochrome fish body image within a monochrome image of the fish body that is irradiated with monochromatic light, bleeding of the light does not occur to the outline thereof, whereby it is possible to obtain an accurate outline.
- FIG. 1 is a diagram illustrating an overall configuration of a fish species identification device of a first embodiment of the present invention.
- a reference numeral 10 denotes a conveyer as a conveying means that conveys a fish body 20 mounted discretely in an arbitrary posture thereon in an X direction marked with an arrow.
- a transparent conveyer as well as a conveyer of an arbitrary color may be used as the conveyer 10.
- a reference numeral 30 denotes a sensor that detects a leading end in a conveyance direction of the fish body 20 conveyed on the conveyer 10, and for example, an optical sensor may be used.
- a reference numeral 40 denotes a light source as an irradiation means that irradiates a surface of the conveyer 10 with a near-infrared light ⁇ 1 (single wave or multiple waves having a wavelength substantially from 700 nm to 1000 nm) and with a white light ⁇ 2 (light containing most of visible light and having a wavelength substantially from 400 nm to 700 nm), and for example, a configuration may be employed in which a ring-arranged LED lamp for radiating the near-infrared light and a ring-arranged LED lamp for radiating the white light are concentrically combined and a reflecting plate 41 is provided on a backside thereof.
- a reference numeral 50 denotes a CCD camera as an imaging means that images the fish body 20, which is on the conveyer 10 and is irradiated by the light source 40.
- the CCD camera having sensitivity capable of imaging both images of wavelengths ⁇ 1 and ⁇ 2 is employed.
- a reference numeral 60 denotes an identification processing unit including image processing using a computer. It performs fish species identification of the fish body 20 that has been imaged by the CCD camera 50 by capturing and processing a monochrome image of the fish body 20 by radiating the near-infrared light and a color image thereof by radiating the white light.
- a reference numeral 70 denotes an operation unit constituted of a keyboard, a mouse, and the like for inputting an instruction of processing details and the like to the identification processing unit 60.
- a reference numeral 80 denotes a monitor for displaying an operational instruction by the operation unit 70, an identification result by the identification processing unit 60, and the like.
- a reference numeral 61 denotes a first frame memory for storing data of the image imaged by the CCD camera 50 by radiating the near-infrared light (hereinafter, referred to as a "monochrome image G1").
- a reference numeral 62 denotes a second frame memory for storing data of the image imaged by the CCD camera 50 by radiating the white light (hereinafter, referred to as a "color image G2").
- a reference numeral 63 denotes a feature quantity extraction unit that extracts a feature quantity of the fish body 20 by extracting an outline of the fish body 20 from the data of the monochrome image G1 read from the first frame memory 61 and by extracting hue of the fish body 20 from the data of the color image G2 read from the second frame memory 62.
- a reference numeral 64 denotes a checking processing unit. By comparing and checking the feature quantity of the fish body 20, which has been detected by the feature quantity extraction unit 63, with a feature quantity of each of a plurality of registered fish bodies registered in a registration unit 65, it identifies a registered fish body having the highest matching degree therewith among the feature quantities of the registered fishbodies. From the checking processing unit 64, afishspecies code indicating the registered fish body that has been identified through this checking is output.
- FIG. 3 a time chart of operation is illustrated.
- a conveyance distance of the conveyer 10, which conveys the fish body 20 at a constant speed, is managed by an encoder (not illustrated).
- the light source 40 irradiates the fish body 20 on the conveyer 10 with the near-infrared light ⁇ 1 at the time when the count value advances to C2 from the time of detection.
- the CCD camera 50 opens a shutter and images the monochrome image G1 of the fish body 20.
- the count value of the encoder becomes C4 irradiation with the near-infrared light ⁇ 1 by the light source 23 is ended.
- the light source 40 irradiates the same fish body 20 on the conveyer 10 with the white light ⁇ 2. Then, when the count value of the encoder becomes C6, the CCD camera 40 opens the shutter and images the color image G2 of the fish body 20. When the count value of the encoder becomes C7, irradiation with the white light ⁇ 2 by the light source 40 is ended.
- Fig. 4 a flowchart of identification processing is illustrated.
- the forefront of the fish body 20 conveyed in the X direction by the conveyer 10 is detected at the time of the count value C1 of the encoder (step S1).
- the fish body 20 is irradiated with the near-infrared light having the wavelength ⁇ 1 from the light source 40, and at the count value C3 after irradiation with the near-infrared light, the monochrome image G1 of the fish body 20 is imaged by the CCD camera 50 (step S2).
- the irradiation with the near-infrared light having the wavelength ⁇ 1 by the light source 40 is ended at the count value C4.
- the data of the monochrome image G1 by the near-infrared light is stored in the first frame memory 61 of the identification processing unit 60.
- the fish body 20 When the count value of the encoder becomes C5, the fish body 20 is irradiated with the white light having the wavelength ⁇ 2 from the light source 40 this time, and at the count value C6 after irradiation with the white light, the color image G2 of the fish body 20 is imaged by the CCD camera 50 (step S3). The irradiation with the white light having the wavelength ⁇ 2 by the light source 40 is ended at the count value C7. Then, the data of the color image G2 by the white light is stored in the second frame memory 62 of the identification processing unit 60.
- the monochrome image G1 stored in the first frame memory 61 becomes an image including a monochrome fish body image 20A such as of black and white as illustrated in Fig. 5(a) , for example.
- the color image G2 stored in the second frame memory 62 becomes an image including a color fish body image 20B having RGB densities (although colors are omitted here) as illustrated in Fig. 5 (b) , for example, and the outline of the color fish body image 20B may become unclear in some cases.
- a measurement region of the color fish body image 20B is detected for extracting the feature quantity of the fish body 20 by overlapping the color fish body image 20B within the second color image G2 on the monochrome fish body image 20A within the first monochrome image G1 illustrated in Fig. 5(a) .
- the color fish body image 20B within the second color image G2 is an image that is moved 20 mm in the conveyance direction from the monochrome fish body image 20A within the first monochrome image G1.
- an angle of view of both of the images 20A and 20B is 320 mm in the X direction of conveyance and the number of horizontal pixels in one screen is 640 pixels, one pixel is equivalent to 0.5 mm.
- a moving distance of 20 mm is equivalent to 40 pixels when converted into the number of pixels.
- a starting point of reading in a horizontal direction thereof is set to a point of a pixel that is shifted by 40 pixels (-40 pixels) from a starting point of reading in the horizontal direction of the first frame memory 61 of the first monochrome fish body image 20A. Accordingly, by overlapping the images read from the first frame memory 61 and the second frame memory 62 in the same timing, the second color fish body image 20B is accurately overlapped on the first monochrome fish body image 20A of the fish body 20, whereby the measurement region of the color fish body image 20B is specified.
- the feature quantity of the fish body 20 is extracted after the measurement region of the color fish body image 20B has been specified as described above.
- This processing is performed through shape extraction of the outline and hue extraction of the color image.
- the shape extraction is performed based on a ratio W/L of a length L and a width W of the outline of the monochrome fish body image 20A. For example, when the ratio W/L is smaller than a predetermined value, it may be determined as a narrow fish such as a saury, and when it is larger, it may be determined as a broad fish such as a mackerel.
- Fig. 6(b) it is also possible to determine whether it is the narrow fish or the broad fish by obtaining a center line 21 from a head to a tail in the monochrome fish body image 20A, by dividing the line 21 into 30, and by measuring widths Wa1 to Wa30 from the line 21 to an edge on a side of a dorsal fin 22 for each of divided regions as well as measuring widths Wb1 to Wb30 from the line 21 to an edge on a side of a pectoral fin 23.
- hue extraction by using the color fish body image 20B within the measurement region that is specified by the outline of the monochrome fish body image 20A, an area of hue of an individual part is measured, or an area of a pattern of a specific part is measured.
- the feature quantity of the fish body 20 is extracted as described above.
- checking processing is performed with the feature quantity of each of the plurality of registered fish bodies registered to the registration unit 65 in advance (step S5).
- the checking processing when the matching degree of the feature quantity of the fish body 20 with the feature quantity of a registered fish body exceeds a threshold set in advance, the fish body 20 is assumed to be the registered fish body and obtains the fish species code of the registered fish body (step S6).
- the fish species code is allocated to a variety of fishes as illustrated in Fig. 7 , for example.
- the fish body which has already been identified its fish species and is conveyed by the conveyer 10, can be classified into the fish species and be sorted into each bucket by the classification device.
- step S4 details of the feature quantity extraction processing (step S4) and the checking processing (S5) in a case where a known mackerel is used as the fish body 20 are described with reference to Figs. 6 (a) and 6(b) .
- the monochrome fish body image 20A is divided into 30 in a lengthwise direction from a head to a tail, a length in a width direction of the monochrome fish body image 20A is measured for each of the divided regions, and a pixel position of a midpoint of the length in the width direction is specified.
- the midpoint obtained in each of the divided regions is connected with each other from one end to the other end, and by smoothing a connected line, the substantially linear line 21 is obtained.
- an upper side region 24 of the line 21 and a lower side region 25 thereof are specified, and each of areas of the upper side region 24 and the lower side region 25 is measured. Based on this measurement result, a side with a larger area is considered to be a dorsal side region and a side with a smaller area is considered to be an abdominal side region.
- hue on the dorsal side region of the fish body 20 is obtained. It is compared with an area ratio of hue on a dorsal side of a mackerel registered to the registration unit 65 in advance as well as an area ratio of an entire area on the dorsal side.
- hue and an area are obtained and are compared with hue and an area registered to the registration unit 65 in advance. In this comparison, it is determined as a blue mackerel when the hue is the registered hue and the area is equal to or greater than the registered area. Even though the hue is the registered hue, when the area is equal to or less than the registered area, it is determined as a common mackerel.
- the fish body 20 of a known fish species By conveying the fish body 20 of a known fish species as a test by the conveyer 10, it is preferred that a feature quantity of the fish species obtained by the feature quantity extraction unit 63 be registered to the registration unit 65 (studying and registering).
- the fish body 20 of the known fish species In screening in which a fish body is identified and classified, in general, the fish body 20 of the known fish species is a target, whereby by conveying the fish body 20 on the conveyer 10 for performing identification operation and by registering the fish species to the registration unit 65 in advance, it is possible to perform the fish species identification with high probability.
- Fig. 8 is a diagram illustrating an overall configuration of a fish species identification device of a second embodiment of the present invention.
- an imaging means two CCD cameras 50A and 50B arranged along a conveyance direction X of a conveyer 10 are used.
- a monochrome image is imaged by the CCD camera 50A with a near-infrared light ⁇ 1
- a color image is imaged by the CCD camera 50B with a white light ⁇ 2.
- the CCD camera 50A may have sensitivity to the near-infrared light ⁇ 1 only, and the CCD camera 50B may have the sensitivity to the white light ⁇ 2 only, whereby it is possible to use cameras optimal for the CCD cameras 50A and 50B.
- 20 msec of time is necessary from imaging of the first monochrome fish body image 20A to imaging of the second color fish body image 20B, whereby it is necessary to read the second color fish body image 20B from a second frame memory 62 by shifting pixels equivalent to a moving distance during the time (40 pixels), or 20 mm, for overlapping the second color fish body image 20B on the first monochrome fish body image 20A.
- the monochrome fish body image 20A written in a frame memory 61 and the color fish body image 20B written in the frame 62 are read and overlapped on each other in the same timing, whereby specification of a measurement region becomes possible in feature quantity extraction processing of step S4 in Fig. 4 .
- a fish body 20 is irradiated with the near-infrared light ⁇ 1 for imaging a monochrome fish body image 20A; however, it is not limited to the near-infrared light ⁇ 1, and it may also be an infrared light or another monochromatic light (light having a single wavelength or light having a plurality of approximating wavelengths).
- a conveyer 10 a color pattern thereof is not particularly limited as long as it is an ordinary conveyer.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Software Systems (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- Computing Systems (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Human Computer Interaction (AREA)
- Processing Of Meat And Fish (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Image Processing (AREA)
- Food Science & Technology (AREA)
- Image Analysis (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Geometry (AREA)
- Medicinal Chemistry (AREA)
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/JP2014/082665 WO2016092646A1 (ja) | 2014-12-10 | 2014-12-10 | 魚種判別装置および魚種判別方法 |
Publications (1)
Publication Number | Publication Date |
---|---|
EP3239925A1 true EP3239925A1 (en) | 2017-11-01 |
Family
ID=53277275
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
EP14875888.1A Withdrawn EP3239925A1 (en) | 2014-12-10 | 2014-12-10 | Fish type determination device and fish type determination method |
Country Status (4)
Country | Link |
---|---|
EP (1) | EP3239925A1 (ja) |
JP (1) | JP5713376B1 (ja) |
KR (1) | KR101656635B1 (ja) |
WO (1) | WO2016092646A1 (ja) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2020016501A (ja) * | 2018-07-24 | 2020-01-30 | 日本電気株式会社 | 計測装置、計測システム、計測方法およびコンピュータプログラム |
CN111223137A (zh) * | 2018-11-26 | 2020-06-02 | 重庆小雨点小额贷款有限公司 | 一种养殖产品的数量确定方法、装置、终端及存储介质 |
CN111357800A (zh) * | 2020-03-26 | 2020-07-03 | 江苏省海洋水产研究所 | 渔获识别分离装置 |
Families Citing this family (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2018037638A1 (ja) * | 2016-08-26 | 2018-03-01 | Necソリューションイノベータ株式会社 | 生存魚管理システム |
CN106417104A (zh) * | 2016-08-31 | 2017-02-22 | 北京农业信息技术研究中心 | 养殖鱼类游动方向测定系统和方法 |
CN106779042B (zh) * | 2016-12-27 | 2023-08-04 | 北京农业信息技术研究中心 | 一种水产养殖鱼群聚集指数计算装置及计算方法 |
JP6917762B2 (ja) * | 2017-05-09 | 2021-08-11 | 株式会社キーエンス | 画像検査装置 |
JP6890789B2 (ja) * | 2018-02-05 | 2021-06-18 | 株式会社シー・テック | 魚選別装置、魚選別方法、魚種類推定装置、及び、魚種類推定方法 |
JP7001145B2 (ja) * | 2018-03-09 | 2022-01-19 | 日本電気株式会社 | 情報処理装置、物体計測システム、物体計測方法およびコンピュータプログラム |
JP7126858B2 (ja) * | 2018-05-18 | 2022-08-29 | 古野電気株式会社 | 魚種推定システム、魚種推定方法、及びプログラム |
JP6780682B2 (ja) * | 2018-09-20 | 2020-11-04 | 日本電気株式会社 | 情報取得システム、制御装置及び情報取得方法 |
JP2020051982A (ja) * | 2018-09-28 | 2020-04-02 | Jfeテクノリサーチ株式会社 | 画像検査装置及び検査モデル構築システム |
KR102191062B1 (ko) | 2019-03-14 | 2020-12-14 | 경남과학기술대학교 산학협력단 | 어종 분류를 위한 데이터 적용 방법 |
KR102191064B1 (ko) | 2019-03-14 | 2020-12-14 | 경남과학기술대학교 산학협력단 | 어종 분류를 위한 데이터 적용 시스템 |
KR102295255B1 (ko) | 2020-09-25 | 2021-08-27 | 최진호 | 생태교란 어종의 생태 조사 및 포획시스템 |
WO2022260034A1 (ja) * | 2021-06-09 | 2022-12-15 | 株式会社エヌ・クラフト | 水産物選別システムおよび船 |
CN114821778A (zh) * | 2022-04-22 | 2022-07-29 | 中国农业大学 | 一种水下鱼体姿态动态识别方法及装置 |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP3483039B2 (ja) * | 1993-09-30 | 2004-01-06 | 博文 松尾 | 画像信号による識別装置 |
JP3377697B2 (ja) | 1996-10-08 | 2003-02-17 | 三菱重工業株式会社 | 魚種判別処理装置 |
JP2006170669A (ja) * | 2004-12-13 | 2006-06-29 | Mitsui Mining & Smelting Co Ltd | 青果物の品質検査装置 |
JP5455409B2 (ja) * | 2009-03-30 | 2014-03-26 | 日立造船株式会社 | 異物選別方法および異物選別設備 |
-
2014
- 2014-12-10 WO PCT/JP2014/082665 patent/WO2016092646A1/ja active Application Filing
- 2014-12-10 JP JP2015505355A patent/JP5713376B1/ja active Active
- 2014-12-10 EP EP14875888.1A patent/EP3239925A1/en not_active Withdrawn
- 2014-12-10 KR KR1020157013286A patent/KR101656635B1/ko active IP Right Grant
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2020016501A (ja) * | 2018-07-24 | 2020-01-30 | 日本電気株式会社 | 計測装置、計測システム、計測方法およびコンピュータプログラム |
CN111223137A (zh) * | 2018-11-26 | 2020-06-02 | 重庆小雨点小额贷款有限公司 | 一种养殖产品的数量确定方法、装置、终端及存储介质 |
CN111223137B (zh) * | 2018-11-26 | 2023-06-16 | 重庆小雨点小额贷款有限公司 | 一种养殖产品的数量确定方法、装置、终端及存储介质 |
CN111357800A (zh) * | 2020-03-26 | 2020-07-03 | 江苏省海洋水产研究所 | 渔获识别分离装置 |
CN111357800B (zh) * | 2020-03-26 | 2021-12-14 | 江苏省海洋水产研究所 | 渔获识别分离装置 |
Also Published As
Publication number | Publication date |
---|---|
JPWO2016092646A1 (ja) | 2017-04-27 |
WO2016092646A1 (ja) | 2016-06-16 |
KR20160084810A (ko) | 2016-07-14 |
JP5713376B1 (ja) | 2015-05-07 |
KR101656635B1 (ko) | 2016-09-09 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP3239925A1 (en) | Fish type determination device and fish type determination method | |
JP7391173B2 (ja) | 食品検査補助システム、食品検査補助装置、およびコンピュータプログラム | |
CN108445007B (zh) | 一种基于图像融合的检测方法及其检测装置 | |
TWI313576B (en) | Board inspecting apparatus, its parameter setting method and parameter setting apparatus | |
WO2019151393A1 (ja) | 食品検査システム、食品検査プログラム、食品検査方法および食品生産方法 | |
KR102003781B1 (ko) | 초분광영상화 기법을 이용한 글라스(Glass) 결함 검출 장치 | |
JP6348289B2 (ja) | 検査装置および検査方法 | |
US20180156735A1 (en) | Colour measurement of gemstones | |
JP7146012B2 (ja) | 薬剤検査支援装置、薬剤識別装置、画像処理装置、画像処理方法及びプログラム、記録媒体 | |
TW201019237A (en) | Method and system for item identification | |
Yang et al. | Machine vision system for on-line wholesomeness inspection of poultry carcasses | |
JP2008073158A (ja) | 画像処理装置 | |
JP2018146251A (ja) | 異物検出システム、異物検出方法、及びそのプログラム | |
JP2016090476A (ja) | 異物検出方法 | |
JPH0534281A (ja) | メロンの外観評価装置 | |
JP6623545B2 (ja) | 検査システム、検査方法、プログラムおよび記憶媒体 | |
CN112236668A (zh) | 谷物的光泽测定装置 | |
JP2019203845A (ja) | 汚卵検出装置 | |
EP4130725A1 (en) | Foreign matter inspection device | |
KR20230139166A (ko) | 목재 제품의 불량 검사 방법 | |
JP7136064B2 (ja) | 被検査体の表面検査装置および被検査体の表面検査方法 | |
Heitschmidt et al. | Improved imaging system for fecal detection | |
JP2017026571A (ja) | 画像処理システム、画像処理装置およびプログラム | |
Sandnes | An automatic two-hand 2D: 4D finger-ratio measurement algorithm for flatbed scanned images | |
Carstensen et al. | Creating surface chemistry maps using multispectral vision technology |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PUAI | Public reference made under article 153(3) epc to a published international application that has entered the european phase |
Free format text: ORIGINAL CODE: 0009012 |
|
17P | Request for examination filed |
Effective date: 20150709 |
|
AK | Designated contracting states |
Kind code of ref document: A1 Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR |
|
AX | Request for extension of the european patent |
Extension state: BA ME |
|
DAX | Request for extension of the european patent (deleted) | ||
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: THE APPLICATION IS DEEMED TO BE WITHDRAWN |
|
18D | Application deemed to be withdrawn |
Effective date: 20180703 |